This proposal is for support of research to explore a biochemical method of cancer detection based on the quantitation of serum or plasma levels of a lipid-soluble sialic acid (LASA) fraction which is consistently and markedly elevated in tumor-bearing animal models and in human cancer patients. Specific objectives are: 1) To maximize reproducibility and sensitivity of the assay procedure; 2) to correlate LASA levels with tumorigenesis in a broad range of animal models and with several tumor types to gain information on the generality of the procedure as a diagnostic tool; 3) to isolate and identify chemically the chemical constituents of the LASA fraction which are elevated as a result of tumor burden in order to provide basic information necessary to understand the underlying biological mechanisms and to increase specificity and sensitivity of the assay procedure.